Abstract
Objectives
Right atrial (RA) strain is increasingly recognized as a significant predictor of adverse events in patients with various cardiovascular conditions. However, the prognostic value of RA strain in patients with left ventricular noncompaction (LVNC) is unclear. The objective of this study was to evaluate the prognostic significance of RA strain derived from cardiac magnetic resonance feature tracking (CMR-FT) in patients with LVNC.
Methods
394 LVNC patients who underwent CMR at 4 Chinese medical facilities from September 2014 to July 2023 were retrospectively and consecutively included in total. RA strain parameters were obtained using CMR-FT. Major adverse cardiac events (MACEs) were assessed, and all patients were followed up.
Results
156 patients (39.6%) experienced MACEs during a median follow-up of 34 months. At univariable analysis, RA conduit strain was associated with MACE (hazard ratio [HR] 0.88 [95% CI 0.85–0.91]; P < 0.001). RA conduit strain maintained an independent predictor of MACE in a multivariate model that included left ventricular ejection fraction (LVEF) and late gadolinium enhancement (LGE) (HR 0.89 [95% CI 0.85–0.93]; p < 0.001). Furthermore, adding RA conduit strain to the multivariate model greatly enhanced the prognostic role of endpoint events (C-statistic improvement: 0.766–0.871, Delong test: p < 0.001). Net reclassification index (NRI) (0.201, p < 0.05) and integrated discrimination improvement (IDI) (0.038, p < 0.05) also showed the same trend.
Conclusion
CMR-FT derived RA conduit strain is a potent independent indicator of major adverse cardiac events in left ventricular noncompaction patients. In addition, RA conduit strain can provide additional prognostic value over the multivariable baseline clinical model.
Graphical abstract
Supplementary Information
The online version contains supplementary material available at 10.1186/s12967-025-07166-y.
Keywords: Left ventricular noncompaction, Cardiac magnetic resonance feature tracking, Right atrial strain, Prognosis
Introduction
Left ventricular noncompaction (LVNC) is a rare, heterogeneous condition distinguished by a myocardium with two distinct layers with noncompacted and compacted layers, pronounced myocardial trabeculations, deep inter-trabecular recesses (noncompacted layer) connected to the LV chamber, and a thin compacted epicardial layer [1, 2]. It can lead to serious consequences, including ventricular arrhythmias (VAs), heart failure (HF), cardiac-related death, and systemic embolisms [3]. Although it has been nearly 30 years since it was first described as a clinical entity, the prognosis for patients remains uncertain [4]. Sedaghat-Hamedani F et al. carried out a study and found LVNC patients exhibited a greater frequency of cardiovascular incidents and seemed to have a poorer outcome than those with dilated cardiomyopathy (DCM) who were age- and sex- matched [5]. Despite being the primary prognostic indicators for LVNC patients, left ventricular ejection fraction (LVEF) and late gadolinium enhancement (LGE) sometimes exhibit a poor correlation with clinical outcomes [6, 7]. Consequently, risk categorization for LVNC is ambiguous and challenging; better tools for risk assessment need to be explored.
The right atrium is essential for maintaining proper cardiac hemodynamics and regulating ventricular filling. It performs three basic functions throughout the cardiac cycle. (1) acting as a reservoir, holding the return of systemic venous blood during ventricular systole and atrial diastole; (2) acting as a conduit, permitting blood to enter the ventricular cavity passively in the early stages of ventricular diastole; and (3) acting as a booster pump, pumping blood into the right ventricle actively during late ventricular diastole [8, 9]. Although physiologically significant, information about the prognostic significance of right atrial (RA) function in LVNC patients is lacking.
Being the non-invasive benchmark for measuring the morphology and function of ventricles, cardiac magnetic resonance (CMR) could offer insights into prognosis, tissue properties, morphology, and function [10–12]. Cardiac magnetic resonance feature tracking (CMR-FT) is an innovative approach that assesses myocardial deformation by tracking the movement of tissue voxels [13], which has been shown to be a sensitive marker for evaluating aberrant cardiac deformation [14]. Moreover, RA strain derived from this technique holds potential predictive significance in several cardiovascular diseases, for instance, dilated cardiomyopathy (DCM), arrhythmogenic right ventricular cardiomyopathy (ARVC), and heart failure [15–17]. Nevertheless, the prognostic significance of RA strain in LVNC patients hasn’t been thoroughly explored in large, multicenter research. Therefore, this research sought to assess the prognostic significance of CMR-FT RA strain in a group of LVNC patients from different centers.
Methods
Subjects
A retrospective analysis was carried out on continuous patients who had CMR at 4 Chinese medical facilities from September 2014 to July 2023. These patients met the Jenni criteria [18] for LVNC based on 2-dimensional transthoracic echocardiography (TTE), as well as satisfied the Petersen [19] and Jacquier [20] criteria by CMR. The exclusion criteria were the following: (1) congenital heart diseases; (2) ischemic heart disease (detailed definitions of ischemic heart disease are available in Appendix Supplementary Materials); (3) moderate to severe valvular diseases; (4) atrial fibrillation; (5) the presence of other genetic or acquired cardiomyopathy (such as dilated cardiomyopathy, hypertrophic cardiomyopathy, hypertensive heart disease, etc.); (6) specific groups of people (< 18 years, pregnant women and athletes); (7) contraindications for CMR. Clinical information was gathered for every subject.
CMR protocol
CMR was performed on 3.0-T scanners (MAGNETOM Prisma, MAGNETOM skyra, MAGNETOM Verio, vida, Siemens Healthcare; Ingenia CX, Elition, Philips; Discovery MR750, General Electric Healthcare) at four centers. All four participating medical centers followed a standardized CMR acquisition protocol [21] in accordance with Society for Cardiovascular Magnetic Resonance (SCMR) guidelines. CMR images were obtained using breath-holding and electrocardiographic gating. Steady-state free precession (SSFP) [22] cines and LGE utilizing phase-sensitive inversion recovery (PSIR) [23] technology were included. SSFP cine images were obtained in standard two-, three-, and four-chamber long-axis as well as short-axis cine stacks covering the ventricles. LGE imaging was conducted 10–15 min following the injection of 0.2 mmol/kg gadopentetate dimeglumine (Bei Lu Pharmaceutical Co. Ltd.). The detailed parameters are presented in Appendix Supplementary Materials.
Image analysis
Post-processing was conducted centrally using the same commercially available software CVI42 (version 5.17.0). Two seasoned CMR physicians, unaware of clinical details, independently assessed images with the CVI42 (version 5.17.0). The volume, mass, and ejection fraction of the ventricles were measured on SSFP cine images. LV trabeculations, defined as any irregularities in the contour of the endocardial wall observed at end-diastole and different from chordae and papillary muscles, were quantified on long-axis images following the approach proposed by Petersen et al.19. LGE was assessed by whether it was present or absent. Manual outlining of the left atrial (LA) endocardial contours was performed in the two- and four-chamber views, while the RA endocardial contours were traced manually in the four-chamber view. The emptying fraction was calculated using the single area-length method, while the biatrial volume was determined with the biplane area-length method. The parameters of biventricular and LA strain, including global radial (GRS), longitudinal (GLS), and circumferential (GCS) strain for the LV and RV, along with LA reservoir, conduit, and booster strain, were measured utilizing CMR-FT referring to previous methods [13, 24]. The epicardial and endocardial boundaries of RA were delineated manually in the four-chamber images utilizing a point-and-click method at the minimal RA volume following atrial contraction [24]. Prior to implementing an automated tracking technique, the RA appendage and superior vena cava were carefully excluded from the RA endocardial outlines. RA strain parameters were obtained employing CMR-FT.
Follow-up
All MACEs were independently reviewed and confirmed by a blinded clinical endpoint committee consisting of two seasoned CMR physicians from different centers. In cases of disagreement, a third senior cardiologist adjudicated the final outcome. To ensure consistency, investigators from each center underwent centralized training, and standardized case report forms were used for event reporting. Two seasoned CMR physicians, unaware of the CMR results, independently determined MACEs and performed the follow-up. Follow-up data were obtained via hospital records, telephone interviews, and clinical visits every six months following the CMR examination. MACEs were assessed up until January 1, 2024, after which the dataset was finalized and locked. MACEs were described as follows: admission for heart failure, VAs, cardiac-related mortality, heart transplantation, appropriate implantable cardioverter defibrillator (ICD) treatment, and systemic embolisms. The specific definitions of VAs, appropriate implantable cardioverter defibrillator (ICD) treatment, and systemic embolisms can be found in Appendix Supplementary materials. Only the first event was taken into account when a patient had multiple events.
Statistical analysis
Statistical analysis was performed with SPSS (version 27.0), GraphPad Prism (version 9.0.0), MedCalc (version 15.8), and R (version 4.0.4). A two-tailed p-value less than 0.05 was deemed statistically significant. The normality of continuous variables was evaluated employing the Shapiro-Wilk test. Continuous variables are shown as medians with interquartile range (IQR) or means ± standard deviation (SD) and categorical variables are expressed as frequencies and percentages. Normally distributed continuous variables, categorical variables, and skewed distributed continuous variables were compared using Student’s t-tests, χ2 tests, and Mann-Whitney U tests, respectively. Missing data (< 2% per variable) were imputed under a missing -at -random assumption using multiple imputation by chained equations. The receiver operating characteristic (ROC) curve was used to evaluate the performance of strain parameters in predicting MACE. The optimal cutoff point for detecting MACE using the RA strain parameter was determined based on the Youden index. The Delong test was employed to compare the prognostic capabilities of RA strain with those of LV, LA, and RV strains. Using the RA strain parameter as a binary classification variable, the Kaplan-Meier survival estimate was calculated based on the optimal cutoff value, and the log-rank test was conducted to assess differences in survival curves. Subgroup analysis was performed to further evaluate the prognostic value of RA conduit strain through Kaplan-Meier survival curves. Cox proportional hazards regression was employed to assess the univariate relationships between pertinent covariates and MACE, thereby identifying risk variables. The Schoenfeld residuals test and visual examination of the log-log survival curves were used to assess the proportional hazards assumption. To verify whether RA conduit strain remains strongly associated with prognosis after eliminating the influence of confounding variables, we constructed the following multivariable Cox proportional hazards models. Model 1 was adjusted for age, male, BMI, LVNC family history, ARNI, and comorbidities (e.g., hypertension, diabetes, and hypercholesterolemia). Model 2 including NYHA functional class, ARNI, LVEF, LGE, LVGLS, RAEF passive, RA reservoir strain, and RA conduit strain was adjusted for univariably associated variables in the backward selection model. A multivariable baseline clinical model was established to examine the independent relationship between RA conduit strain and MACE. Subsequently, RA conduit strain was integrated into the model to evaluate its additional predictive significance. The predictive performance of the multivariable model was evaluated utilizing integrated discrimination improvement (IDI), C-statistic, a likelihood ratio test, and net reclassification index (NRI) for MACE in LVNC patients prior to and subsequent to the incorporation of RA conduit strain. Moreover, the 3- year calibration curves were generated to confirm the prediction ability of the established models. Differences were compared employing the Delong test. The intra-observer and inter-observer reproducibility and variability for RA strain were tested using intraclass correlation coefficients (ICC) and Bland-Altman analysis.
Results
Patient characteristics and CMR parameters
Initially, 560 patients met the inclusion criteria, while 166 patients were not included: 126 patients had concurrent other heart diseases (23 dilated cardiomyopathy, 43 ischemic heart disease, 13 hypertrophic cardiomyopathy, 5 hypertensive heart disease, 25 myocarditis, 10 atrial fibrillation, 4 cardiac amyloidosis, and 3 congenital heart diseases); 28 lacked follow-up, and 12 had poor-quality CMR images. Thus, the sample comprised 394 patients finally (Fig. 1). The key baseline characteristics categorized by MACE were provided in Table 1. 156 patients (39.6%, 43 ± 15 years, 65% male) encountered MACEs, while 238 patients (60.4%, 41 ± 16 years, 63% male) did not. Patients who experienced MACEs had a higher proportion classified in New York Heart Association (NYHA) functional class > II (104, 67%).
Fig. 1.
Flowchart shows patient inclusion. LVNC = left ventricular noncompaction
Table 1.
Clinical and CMR characteristics according to major adverse cardiac events
| Overall (n = 394) |
No MACE | MACE | P values | |
|---|---|---|---|---|
| (n = 238) | (n = 156) | |||
| Clinical data | ||||
| Age, year | 42 ± 16 | 41 ± 16 | 43 ± 15 | 0.254 |
| Male (n, %) | 251 (64) | 149 (63) | 102 (65) | 0.575 |
| BMI, kg/m2 | 25 ± 3 | 25 ± 3 | 25 ± 3 | 0.536 |
| Hypertension (n, %) | 93 (24) | 60 (25) | 33 (21) | 0.354 |
| Diabetes (n, %) | 74 (19) | 43 (18) | 31 (20) | 0.654 |
| Hypercholesterolemia (n, %) | 68 (17) | 42 (18) | 26 (17) | 0.801 |
| LVNC family history (n, %) | 36 (9) | 17 (7) | 19 (12) | 0.090 |
| Smoker (n, %) | 116 (29) | 69 (29) | 47 (30) | 0.809 |
| Abnormal ECG (n, %) | 188 (48) | 116 (49) | 72 (46) | 0.616 |
| NYHA functional class > II (n, %) | 235 (60) | 131 (55) | 104 (67) | 0.022 |
| Therapy | ||||
| ACEI/ARB (n, %) | 115 (29) | 73 (31) | 42 (27) | 0.424 |
| ARNI (n, %) | 119 (30) | 49 (21) | 70 (45) | < 0.001 |
| Beta-receptor blocker (n, %) | 109 (28) | 68 (29) | 41 (26) | 0.620 |
| MRAs (n, %) | 164 (42) | 90 (38) | 74 (47) | 0.059 |
| Ivabradine (n, %) | 142 (36) | 88 (37) | 54 (35) | 0.634 |
| Diuretic (n, %) | 181 (46) | 115 (48) | 66 (42) | 0.242 |
| Calcium channel blockers(n,%) | 50 (13) | 31 (13) | 19 (12) | 0.805 |
| Antiarrhythmic(n,%) | 71 (18) | 45 (19) | 26 (17) | 0.572 |
| OAC (n,%) | 93 (24) | 55 (23) | 38 (24) | 0.775 |
| CMR parameters | ||||
| LVEF (%) | 41 ± 17 | 46 ± 15 | 33 ± 15 | < 0.001 |
| LVEDV index (ml/m2) | 100 ± 42 | 97 ± 44 | 104 ± 40 | 0.073 |
| LVESV index (ml/m2) | 60 ± 42 | 55 ± 42 | 68 ± 39 | 0.004 |
| LVSV index (ml/m2) | 40 ± 15 | 41 ± 13 | 37 ± 19 | 0.015 |
| LV Mass index (g/m2) | 68 ± 27 | 67 ± 27 | 70 ± 27 | 0.363 |
| LGE present (n, %) | 219 (56) | 101 (42) | 118 (76) | < 0.001 |
| LVGRS (%) | 22 ± 10 | 23 ± 10 | 21 ± 11 | 0.206 |
| LVGCS (%) | − 14 ± 6 | − 14 ± 6 | − 13 ± 6 | 0.040 |
| LVGLS (%) | − 12 ± 5 | − 12 ± 5 | − 11 ± 5 | < 0.001 |
| NC/C ratio | 3.3 ± 1.0 | 3.2 ± 1.1 | 3.4 ± 1.0 | 0.065 |
| RVEF (%) | 42 ± 16 | 44 ± 15 | 38 ± 16 | < 0.001 |
| RVEDV index (ml/m2) | 80 ± 24 | 80 ± 22 | 80 ± 27 | 0.839 |
| RVESV index (ml/m2) | 48 ± 22 | 46 ± 20 | 50 ± 25 | 0.046 |
| RVSV index (ml/m2) | 33 ± 14 | 35 ± 14 | 30 ± 14 | < 0.001 |
| RVGRS (%) | 22 ± 9 | 22 ± 9 | 21 ± 9 | 0.087 |
| RVGCS (%) | − 13 ± 5 | − 13 ± 5 | − 12 ± 5 | 0.017 |
| RVGLS (%) | − 15 ± 6 | − 16 ± 6 | − 14 ± 6 | 0.008 |
| LAVI max (ml/m2) | 40 ± 18 | 39 ± 16 | 42 ± 20 | 0.111 |
| LAVI pac (ml/m2) | 30 ± 16 | 29 ± 15 | 32 ± 18 | 0.063 |
| LAVI min (ml/m2) | 20 ± 15 | 19 ± 14 | 22 ± 16 | 0.040 |
| LAEF total (%) | 53 ± 17 | 54 ± 16 | 50 ± 18 | 0.037 |
| LAEF passive (%) | 27 ± 11 | 28 ± 10 | 26 ± 11 | 0.074 |
| LAEF active (%) | 37 ± 16 | 38 ± 15 | 35 ± 17 | 0.075 |
| LA reservoir strain (%) | 29 ± 13 | 31 ± 11 | 27 ± 15 | < 0.001 |
| LA conduit strain (%) | 17 ± 8 | 18 ± 8 | 15 ± 10 | 0.004 |
| LA booster strain (%) | 12 ± 6 | 13 ± 6 | 11 ± 6 | 0.001 |
| RAVI max (ml/m2) | 34 ± 14 | 34 ± 14 | 34 ± 14 | 0.602 |
| RAVI pac (mL/m2) | 24 ± 12 | 25 ± 12 | 24 ± 12 | 0.394 |
| RAVI min (mL/m2) | 16 ± 11 | 16 ± 11 | 15 ± 11 | 0.413 |
| RAEF total (%) | 54 ± 16 | 53 ± 16 | 56 ± 16 | 0.075 |
| RAEF passive (%) | 28 ± 15 | 27 ± 14 | 31 ± 16 | 0.005 |
| RAEF active (%) | 37 ± 15 | 37 ± 15 | 38 ± 15 | 0.784 |
| RA reservoir strain (%) | 34 ± 10 | 37 ± 9 | 30 ± 11 | < 0.001 |
| RA conduit strain (%) | 15 ± 5 | 17 ± 4 | 12 ± 4 | < 0.001 |
| RA booster strain (%) | 19 ± 10 | 19 ± 10 | 18 ± 11 | 0.447 |
MACE, major adverse cardiac events; BMI, body mass index; LVNC, left ventricular noncompaction; ECG, electrocardiogram; NYHA, New York Heart Association; ACEI, angiotensin-conversion enzyme inhibitor; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor-neprilysin inhibitor; MRA, mineralocorticoid receptor antagonist; OAC, oral anticoagulation; CMR, cardiac magnetic resonance; LV, left ventricular; LVEF, LV ejection fraction; LVEDV index, LV end-diastolic volume index; LVESV index, LV end-systolic volume index; LVSV index, LV stroke volume index; LGE, late gadolinium enhancement; LVGRS, LV global radial strain; LVGCS, LV global circumferential strain; LVGLS, LV global longitudinal strain; NC/C, noncompacted/compacted; RV, right ventricular; RVEF, RV ejection fraction; RVEDV index, RV end-diastolic volume index; RVESV index, RV end-systolic volume index; RVSV index, RV stroke volume index; RVGRS, RV global radial strain; RVGCS, RV global circumferential strain; RVGLS, RV global longitudinal strain; LA, left atrial; LAVI max: maximal LA volume index; LAVI pac: LA volume index prior to atrial contraction; LAVI min: minimal LA volume index; LAEF, LA ejection fraction; RA, right atrial; RAVI max, maximal RA volume index; RAVI pac, RA volume index prior to atrial contraction; RAVI min, minimal RA volume index
LVEF was 41 ± 17%, and 219 (56%) patients exhibited LGE in this study. The MACEs group exhibited lower LVEF, along with a higher occurrence of LGE (all p < 0.001). Furthermore, RA reservoir strain (30 ± 11%) and RA conduit strain (12 ± 4%) were lower in MACEs patients (all p < 0.001) (Table 1). A comparison of CMR characteristics between patients with and without MACEs was presented in Table 1. Figure 2 illustrated instances of RA strain measurements in LVNC patients.
Fig. 2.
Examples of RA strain measurements in two patients with LVNC. a, b Four-chamber CMR scan (a) and RA strain curve (b) in a 56-year-old male patient with RA conduit strain of less than 14.3% who experienced cardiac death 50 months after CMR. c, d Four-chamber CMR scan (c) and RA strain curve (d) in a 37-year-old male patient with RA conduit strain of greater than 14.3% who did not experience MACEs (heart failures needing for cardiac therapy; ventricular arrhythmia requiring therapy, such as cardioverter-defibrillator shock; cardiac death). RA CMR-FT analysis was performed in the four-chamber views (a, c). RA strain values were obtained according to the RA strain curves (b, d). RA, right atrial; LVNC, left ventricular noncompaction; MACE, major adverse cardiac events; CMR-FT, cardiac magnetic resonance feature tracking
Associations of clinical variables and CMR parameters with MACE
Over a median follow-up period of 34 months ([IQR] 19 to 48; range: 6–108), 156 MACEs were recorded, which included 85 hospitalizations due to heart failure, 48 VAs, 2 cardiac-related mortalities, 1 heart transplantation, 15 systemic embolisms (1 peripheral artery embolism, 9 embolic strokes, 5 transient ischemic attacks), and 5 appropriate ICD therapies. Table 2 showed the univariate correlations of clinical variables and CMR indicators with MACE. LVEF, RVEF, NYHA functional class > II, LA reservoir strain, RA reservoir strain, and LGE remained associated with MACE (p < 0.05). Besides, RA conduit strain was a notable univariate predictor of MACE (HR 0.88; 95% CI 0.85–0.91; p < 0.001) (Table 2).
Table 2.
Univariate association of clinical and CMR characteristics with maces
| Parameters | HR | 95% CI | P values |
|---|---|---|---|
| Clinical data | |||
| Age, year | 1.00 | 0.99–1.01 | 0.857 |
| Male | 0.83 | 0.60–1.16 | 0.277 |
| BMI, kg/m2 | 1.01 | 0.95–1.06 | 0.837 |
| LVNC family history | 1.16 | 0.71–1.87 | 0.559 |
| Hypertension | 0.87 | 0.59–1.28 | 0.475 |
| Diabetes | 1.01 | 0.68–1.49 | 0.980 |
| Hypercholesterolemia | 0.76 | 0.50–1.16 | 0.209 |
| Smoker | 1.17 | 0.83–1.65 | 0.372 |
| Abnormal ECG | 0.86 | 0.62–1.17 | 0.335 |
| NYHA functional class > II | 1.40 | 1.00-1.96 | 0.049 |
| ARNI | 1.57 | 1.15–2.16 | 0.005 |
| CMR parameters | |||
| LVEF (%) | 0.97 | 0.96–0.98 | < 0.001 |
| RVEF (%) | 0.99 | 0.98-1.00 | 0.014 |
| LGE present | 2.32 | 1.61–3.35 | < 0.001 |
| NC/C ratio | 1.03 | 0.89–1.19 | 0.714 |
| LVGLS | 1.04 | 1.01–1.07 | 0.014 |
| RVGLS | 1.01 | 0.99–1.04 | 0.398 |
| LA reservoir strain (%) | 0.98 | 0.97-1.00 | 0.010 |
| RAVI max (ml/m2) | 1.00 | 0.99–1.01 | 0.703 |
| RAVI pac (mL/m2) | 1.00 | 0.98–1.01 | 0.510 |
| RAVI min (mL/m2) | 1.00 | 0.98–1.01 | 0.541 |
| RAEF total (%) | 1.01 | 1.00-1.01 | 0.334 |
| RAEF passive (%) | 1.01 | 1.00-1.02 | 0.019 |
| RAEF active (%) | 1.00 | 0.99–1.01 | 0.652 |
| RA reservoir strain (%) | 0.97 | 0.96–0.99 | 0.001 |
| RA conduit strain (%) | 0.88 | 0.85–0.91 | < 0.001 |
| RA booster strain (%) | 1.00 | 0.98–1.01 | 0.703 |
Abbreviations as in Table 1
After adjustment for age, male, BMI, LVNC family history, ARNI, and comorbidities (e.g., hypertension, diabetes, and hypercholesterolemia), RA conduit strain remained independently associated with MACE (Model 1; Table 3). Moreover, in the backward selection model, RA conduit strain also retained a significant association with MACE (Model 2; Table 3).
Table 3.
Multivariable association of clinical and CMR characteristics with maces
| Parameters | Model 1 HR* P values |
Model2 HR* P values |
Model 3 HR* P values |
Model 4 HR* P values |
||||
|---|---|---|---|---|---|---|---|---|
| Age, year | 0.99 (0.98-1.00) | 0.182 | ||||||
| Male | 0.84 (0.59–1.20) | 0.344 | ||||||
| BMI, kg/m2 | 0.96 (0.90–1.02) | 0.177 | ||||||
| LVNC family history | 1.54 (0.93–2.56) | 0.095 | ||||||
| Hypertension | 0.91 (0.57–1.45) | 0.698 | ||||||
| Diabetes | 1.03 (0.68–1.56) | 0.888 | ||||||
| Hypercholesterolemia | 0.91 (0.53–1.54) | 0.716 | ||||||
| NYHA functional class > II | 1.70 (1.18–2.44) | 0.004 | 1.70 (1.18–2.44) | 0.004 | 1.45 (1.03–2.04) | 0.033 | 1.54 (1.09–2.19) | 0.015 |
| ARNI | 0.62 (0.37–1.04) | 0.070 | 0.66 (0.42–1.03) | 0.064 | ||||
| LVEF (%) | 0.97 (0.96–0.99) | 0.001 | 0.98 (0.97–0.99) | 0.007 | 0.98 (0.97–0.99) | < 0.001 | 0.99 (0.98-1.00) | 0.025 |
| RVEF (%) | 1.00 (0.98–1.01) | 0.670 | 1.00 (0.99–1.01) | 0.675 | 1.00 (0.99–1.01) | 0.999 | ||
| LGE present | 1.87 (1.26–2.76) | 0.002 | 1.91 (1.30–2.81) | 0.001 | 2.05 (1.41-3.00) | < 0.001 | 1.93 (1.32–2.82) | < 0.001 |
| LVGLS | 1.04 (1.00-1.07) | 0.043 | ||||||
| LA reservoir strain (%) | 0.99 (0.98-1.00) | 0.144 | 0.99 (0.98-1.00) | 0.062 | 0.99 (0.98–1.01) | 0.291 | ||
| RAEF passive (%) | 1.01 (1.00-1.02) | 0.096 | ||||||
| RA reservoir strain (%) | 0.98 (0.97-1.00) | 0.014 | ||||||
| RA conduit strain (%) | 0.89 (0.85–0.93) | < 0.001 | 0.89 (0.86–0.93) | < 0.001 | 0.89 (0.85–0.93) | < 0.001 | ||
Abbreviations as in Table 1
* Data in parentheses are 95% CIs
Model 1 adjusted for age, male, BMI, LVNC family history, NYHA functional class, ARNI, LVEF, RVEF, LGE, LA reservoir strain, and comorbidities (e.g., hypertension, diabetes, and hypercholesterolemia)
Model 2 adjusted for all univariably associated variables in the backward selection model
Model 3 (The multivariable baseline clinical model) including NYHA functional class, LVEF, RVEF, LGE, and LA reservoir strain
Model 4 = Model 3 (The multivariable baseline clinical model) + RA conduit strain
Analysis of the multivariable model and subgroups
A multivariate model was developed according to the subsequent principles: (1) a significant association with MACE in the univariate analysis with a p-value < 0.001; (2) reflecting healthcare decisions guided by clinical practice and contemporary guidelines [2, 25, 26]. The multivariable model included LVEF, RVEF, LA reservoir strain, NYHA functional class >II, and LGE. The area under the curve (AUC) of RA reservoir strain in predicting MACE is 0.682 (95% CI 0.633–0.727, p < 0.001). The AUC of RA conduit strain for MACE is 0.855 (95% CI 0.816–0.888, p < 0.001) (Fig. 3). Therefore, we incorporated RA conduit strain into the multivariate model. Upon integration into the multivariable model, RA conduit strain was still independently linked to MACE (HR 0.89; 95% CI 0.85–0.93; p < 0.001) (Table 3). Kaplan-Meier survival analysis revealed that patients with “low” RA conduit strain (below the optimal cutoff value of 14.3%) demonstrated a reduced survival rate free from MACE (log-rank, p < 0.001) (Fig. 4). Moreover, in the subgroup analysis, similar results were observed, regardless of LVEF and the existence of LGE (Fig. 5).
Fig. 3.
ROC curve analysis of RA strain compared with LV, LA, and RV strain for the prediction of MACE. ROC, receiver operating characteristic; RA, right atrial; LV, left ventricular; LA, left atrial; RV, right ventricular; GLS, global longitudinal strain
Fig. 4.
A shows the Kaplan-Meier curves for MACE in all patients. B shows the differences in AUC of multivariable models. ROC curve analysis of the multivariate baseline model + RA conduit strain (0.871, 95% CI 0.834 to 0.903) and the multivariate baseline model (0.766, 95% CI 0.721 to 0.807). MACE, major adverse cardiac events; AUC, area under the curve; ROC, receiver operating characteristic; CI, confidence interval
Fig. 5.
Kaplan-Meier curves for MACE in different subgroups. Kaplan-Meier curves of patients with LVEF ≤ 35% (A) and with LVEF > 35% (B) with RA conduit strain < 14.3% versus with RA conduit strain > 14.3% are shown in brown and green, respectively. Kaplan-Meier curves of patients without LGE (C) and with LGE (D) with RA conduit strain < 14.3% versus with RA conduit strain > 14.3% are shown in brown and green, respectively. Log-rank test statistical analysis for RA conduit strain < 14.3% or RA conduit strain > 14.3%
Incremental prognostic value of RA conduit strain
C-statistic improved from (0.766, 95% CI 0.721–0.807) to (0.871, 95% CI 0.834–0.903) (p < 0.001) after the addition of RA conduit strain, suggesting the prognostic value of the model for MACE was significantly improved (Fig. 4) (Table 4). NRI (0.201, p < 0.05) and IDI (0.038, p < 0.05) also showed the same trend (Table 4). The model subsequent to the incorporation of RA conduit strain showed improvement in calibration (likelihood ratio, 80.78) compared with the model prior to the incorporation of RA conduit strain (likelihood ratio, 51.08; P < 0.001). Moreover, the model calibration curves indicate that the model demonstrates enhanced predictive performance with the addition of RA conduit strain, in 3- year follow-up scenarios (Fig. 6).
Table 4.
Model discrimination for MACE after addition of RA conduit strain to the multivariable baseline model
| statistic (95%CI) | p values | |
|---|---|---|
| C-statistic | ||
| Baseline model | 0.766(0.721–0.807) | < 0.001 |
| Baseline model + RA conduit strain | 0.871(0.834–0.903) | < 0.001 |
| Delong test | < 0.001 | |
| NRI | 0.201(0.029–0.328) | < 0.05 |
| IDI | 0.038(0.007–0.105) | < 0.05 |
Baseline model included LVEF, RVEF, NYHA functional class > II, LA reservoir strain, and LGE. NRI, net reclassification index; IDI, integrated discrimination improvement. Abbreviations as in Table 1
Fig. 6.
Incremental value of RA conduit strain over multivariate model. The multivariate model subsequent to the incorporation of RA conduit strain showed improved calibration (likelihood ratio, 80.78) compared with multivariate model that included LVEF, RVEF, LA reservoir strain, NYHA functional class > II, and LGE (likelihood ratio, 51.08; P < 0.001) (A). Calibration of the 2 models in the 3-year follow-up scenarios. Calibration curves for 3-year survival in the multivariate model (B) and calibration curves for 3-year survival in the multivariate + RA conduit strain model (I). Dashed lines along the 45-degree line passing through the point of origin represent perfect calibration models in which predicted probabilities are identical to actual probabilities. X, multivariate model; RA, right atrial; LV, left ventricular; LVEF, LV ejection fraction; RV, right ventricular; RVEF, RV ejection fraction; LA, left atrial; NYHA, New York Heart Association; LGE, late gadolinium enhancement
Comparison of RA strain to LV, LA, and RV strain
The AUC of RA conduit strain (0.855, 95% CI 0.816–0.888) in predicting MACE is higher than that of LA reservoir strain (0.601, 95% CI 0.551–0.650), LVGLS (0.592, 95% CI 0.541–0.641), and RVGLS (0.579, 95% CI 0.528–0.628) (Fig. 3). Further Delong tests indicated that the AUC difference between the RA conduit strain single-variable model and the LA reservoir strain, LVGLS, or RVGLS single-variable models is statistically significant (p < 0.001).
Reproducibility analysis
Reproducibility results are presented in Appendix Supplementary Materials. Both intra-observer and inter-observer reproducibility for assessments of RA reservoir strain (intra: ICC = 0.981, 95% CI[0.967–0.989]; inter: ICC = 0.996, 95% CI [0.993–0.998]), RA conduit strain (intra: ICC = 0.960, 95% CI[0.931–0.977]; inter: ICC = 0.958, 95% CI [0.928–0.976]), and RA booster strain (intra: ICC = 0.998, 95% CI [0.997–0.999]; inter: ICC = 0.991, 95% CI [0.984–0.995]) were excellent(Supplementary Table 2). The variability in both intra-observer and inter-observer measurements of RA reservoir strain, RA conduit strain, and RA booster strain were as follows: RA reservoir strain intra: mean difference − 0.316, 95% CI (− 3.098–2.466);inter: mean difference − 0.254, 95% CI (− 1.582–1.074); RA conduit strain intra: mean difference − 0.428, 95%CI (− 1.827–0.971); inter: mean difference 0.320, 95% CI (− 1.109–1.749); RA booster strain intra: mean difference 0.336, 95%CI (− 0.523–1.195); inter: mean difference 0.578, 95% CI (− 1.371–2.527) (Supplementary Fig. 1).
Discussion
This research broadens the use of RA strain to LVNC patients. To our knowledge, this is the inaugural multicenter evaluation with long-term follow-up of the predictive significance of RA strain in a substantial sample of LVNC patients. The principal conclusions of this research are: (1) As a novel metric for assessing heart function, RA conduit strain is an independent indicator of MACE. (2) In addition to clinical risk indicators such as NYHA functional class > II, LVEF, LA reservoir strain, RVEF, and LGE, RA conduit strain demonstrates a strong predictive correlation with MACE. (3) RA conduit strain provides incremental predictive significance to the multivariable baseline clinical model.
Previous studies have demonstrated that RV dysfunction has prognostic value for patients with LVNC [27–29]. However, as for RA, its function in LVNC patients remains insufficiently understood. Jain S et al. proved that RA conduit and reservoir function independently forecasted mortality in patients with and without heart failure [15]. Zheng J-Y et al. found that RA booster and reservoir strain offered additional predictive significance over CMR predictive and other clinical factors in ARVC [30]. Gao Y et al. identified that CMR-FT RA conduit strain could predict the poor results in DCM patients independently and offer additional prognostic significance in multivariable models [17]. Another study showed that the RA conduit function not only assesses systolic ventricular function after acute myocardial infarction (AMI), but also adds prognostic value, indicating its significant contribution to overall cardiac mechanics [31]. In this research, we discovered that RA conduit strain < 14.3% correlated with an elevated risk of MACE. Through multivariate Cox regression analysis, we further confirmed that RA conduit strain is a strong independent marker of unfavorable prognosis in LVNC patients. Furthermore, we observed no notable difference in RA booster strain between the non-MACE and MACE groups in LVNC patients. Similarly, in patients with pulmonary hypertension, RA conduit and reservoir function was impaired, while the booster function remained preserved [32]. Interestingly, in healthy subjects, with aging, the function of the RA conduit decreases, while the function of the booster pump increases. This similar pattern suggests that as age progresses, the decline in diastolic RV function may lead to enhanced booster pump function and diminished conduit function, which is a compensatory response [33].Therefore, all these collectively indicate that despite the existence of cardiovascular disease, the booster pump function is often preserved.
The RA is crucial in regulating the interaction between the functioning of different chambers and the systemic circulation [9, 34]. However, diastolic and systolic dysfunction in LVNC patients inevitably affects atrial function. In LVNC patients, impaired LV relaxation and elevated LV filling pressures can be transmitted backwards through the pulmonary circulation, increasing RV afterload. Chronically elevated RV pressures may stiffen the RA, reducing its conduit function (early-diastolic reservoir emptying) and leading to elevated RA pressures that further compromise right heart filling and forward cardiac output. Moreover, neurohormonal dysregulation caused by RA fibrosis and stretching, manifested as resistance to natriuretic peptide and excessive defects in the synthesis of atrial natriuretic peptide, leads to sodium retention and volume overload [34]. This drives adverse remodeling in both atria and ventricles. Also, previous studies have demonstrated diffuse atrial fibrosis in non‐compaction cardiomyopathy, which may blunt atrial deformation. Reduced RA conduit strain may therefore reflect subclinical RA myocardial fibrotic remodeling, which is itself linked to arrhythmias, thromboembolism, and progressive heart failure [35]. In addition, myocardial changes in the RA and atrial fibrosis may also affect its electromechanical function. Altered RA mechanics may predispose to atrial arrhythmias (e.g., atrial fibrillation), which are known to worsen outcomes in LVNC. Impaired conduit function could thus identify patients at higher risk of arrhythmia‐related decompensation and thromboembolic events [36, 37]. We also observed that in patients with severely reduced LVEF (< 35%), RA conduit strain remained independently associated with adverse outcomes, suggesting that RA conduit strain may significantly aid in the risk stratification of LVNC patients.
When comparing the prognostic significance of strain in four chambers, we observed that RA conduit strain had the highest predictive ability. We think that because dysfunction of the RV may be a late result of pulmonary circulatory alterations caused by heart failure in LVNC patients, changes in RA strain may help assess right heart function. As the phase function of the RA may respond to early alterations in RV diastolic function and compliance in LVNC patients, particularly RA conduit strain, it may offer more subtle and sensitive prognostic insights for earlier risk stratification of patients.
CMR can offer long-term prognostic insights for LVNC patients, distinguishing those with a high risk of MACE from those with a favorable prognosis [6, 38]. However, the risk stratification of LVNC remains inadequately defined. Therefore, more effective markers are required to be investigated. Our research suggests that RA conduit strain is a dependable predictor of MACE in LVNC patients. LV fibrosis, typically identified through LGE, is a key predictor of adverse prognosis in LVNC patients [6]. We also found this, aligning with prior research findings [39, 40]. Nonetheless, not all LVNC patients exhibit LGE. 219 patients (56%) had LGE, while 175 patients (44%) did not exhibit LGE in our analysis. In LVNC patients lacking LGE, RA conduit strain proved to be an effective predictor of MACE. This discovery indicates that RA conduit strain may enhance risk classification for LVNC patients. Similar to the recommendations for DCM, existing guidelines consider LVEF ≤ 35% as the preferred criterion for primary prevention ICD placement in LVNC patients [41]. Among the 156 MACEs in this study, 102 cases (65%) were with LVEF ≤ 35%, and 54 (35%) were with LVEF >35%. Our findings underscore the constraints of existing predictions exclusively reliant on LVEF ≤ 35%. Moreover, clinically, some adults with LVNC but without left ventricular dysfunction have experienced adverse cardiovascular events. What’s more, the prognosis for LVNC patients with normal LVEF remains unclear [42]. Our study further suggests that RA conduit strain may contribute to earlier risk stratification for LVNC patients. Simultaneously, in regular CMR images, the post-processing software CVI42 (version 5.17.0) is integrated with the Picture Archiving and Communication System, the RA strain can be swiftly and effortlessly calculated with straightforward post-processing, taking about 1–2 min.
This research possesses important clinical implications. In the management and treatment of LVNC patients, the emphasis is primarily on left ventricular function while neglecting right atrial function. Since risk stratification for LVNC patients remains difficult in routine clinical practice, risk stratification should be optimized through more effective indicators. The study results show that in LVNC patients, NYHA functional class > II, LVEF, RVEF, LA reservoir strain, and LGE are clinical variables correlated.
with MACE, aligning with previous findings [25, 26]. Furthermore, RA conduit strain demonstrated an increasing prognostic effect when integrated into the multivariable baseline model, significantly improving risk stratification. Consequently, clinically, RA conduit strain could enhance risk stratification by identifying patients at elevated risk for major adverse cardiac events who might benefit from closer surveillance or earlier therapeutic interventions. Future prospective studies and health economic evaluations are required to confirm whether incorporating RA strain into risk models leads to improved clinical decision-making and patient outcomes.
Limitations
First, its retrospective design introduces an inherent risk of selection bias, and patients who were too ill to undergo CMR or who had incomplete imaging data may have been inadvertently excluded, potentially limiting the generalizability of our findings. Second, although conducted across multiple centers, our cohort was regional and relatively small, with limited representation of non-local ethnic groups. Third, we did not capture data on medication adherence or genetic profiles, which may influence outcomes; future prospective studies with more comprehensive clinical and genomic data are therefore warranted. Fourth, in the absence of an independent external validation cohort, the reproducibility of our results cannot be directly confirmed; multicenter, multinational investigations with broader ethnic representation and external validation samples are needed to establish external validity. Fifth, while software with CMR-FT (e.g., CVI42) is increasingly available in tertiary and academic centers, its adoption may be constrained in less-resourced settings due to limited access and expertise—another factor that could impact the broader applicability of our approach. Finally, RA strain data only comes from the CMR four-chamber long-axis view. Nevertheless, the feasibility and reproducibility of atrial strain assessment in the right atrium have been proven [24].
Conclusion
On balance, this study assessed RA function in LVNC patients and investigated its predictive value for the first time. The research suggested that RA conduit strain is an independent predictor of MACE in LVNC patients, offering additional predictive value over the baseline model. It provides important insights into the organized cardiac mechanics of LVNC patients. Introducing the RA conduit strain as an early sensitive predictor for comprehensive prognostic assessment in LVNC patients is beneficial for improving risk stratification, guiding treatment, and enhancing patient outcomes.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors express thanks to all the editors, reviewers for their kind and valuable suggestion and guidance.
Abbreviations
- LVNC
Left ventricular noncompaction
- VAs
Ventricular arrhythmias
- LVEF
Left ventricular ejection fraction
- HF
Heart failure
- LGE
Late gadolinium enhancement
- DCM
Dilated cardiomyopathy
- RA
Right atrial
- CMR-FT
Cardiac magnetic resonance feature tracking
- TTE
Transthoracic echocardiography
- SSFP
Steady-state free precession
- ARVC
Arrhythmogenic right ventricular cardiomyopathy
- PSIR
Phase-sensitive inversion recovery
- LA
Left atrial
- RVEF
Right ventricular ejection fraction
- GLS
Global longitudinal strain
- SD
Standard deviation
- NYHA
New York Heart Association
- GRS
Global radial strain
- IQR
Interquartile range
- GCS
Global circumferential strain.
- ICD
Iimplantable cardioverter defibrillator.
- ROC
Receiver operating characteristic.
- ICC
Intraclass correlation coefficient.
- NRI
Net reclassification index.
- AUC
Area under the curve.
- IDI
Integrated discrimination improvement.
Author contributions
RX S, Y G, RP W, RZ Z, J W, PX H, and XM W provided substantial contributions to the conception of the work, to the analysis and interpretation of data for the work. RX S and Y G provided substantial contributions to the acquisition of data for the work. JH X, J W, SF Y, CS J, and XS Y provided final approval of the version to be published. All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. All authors declared that they have no conflicts of interest to this work.
Funding
This study was supported by national Natural Science Foundation of China (Grant No. 82271993).
Data availability
The datasets analyzed during the current study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
The ethical committee of Shandong Provincial Hospital approved research procedures.
Consent for publication
Not applicable.
Competing interests
The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Arbustini E, Favalli V, Narula N, Serio A, Grasso M. Left ventricular noncompaction: A distinct genetic cardiomyopathy? J Am Coll Cardiol. 2016;68(9):949–66. [DOI] [PubMed] [Google Scholar]
- 2.Aung N, Doimo S, Ricci F, et al. Prognostic significance of left ventricular noncompaction: systematic review and Meta-Analysis of observational studies. Circulation Cardiovasc Imaging. 2020;13(1):e009712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ross SB, Jones K, Blanch B, et al. A systematic review and meta-analysis of the prevalence of left ventricular non-compaction in adults. Eur Heart J. 2020;41(14):1428–36. [DOI] [PubMed] [Google Scholar]
- 4.Vaidya VR, Lyle M, Miranda WR, et al. Long-Term survival of patients with left ventricular noncompaction. J Am Heart Assoc. 2021;10(2):e015563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sedaghat-Hamedani F, Haas J, Zhu F, et al. Clinical genetics and outcome of left ventricular non-compaction cardiomyopathy. Eur Heart J. 2017;38(46):3449–60. [DOI] [PubMed] [Google Scholar]
- 6.Andreini D, Pontone G, Bogaert J, et al. Long-Term prognostic value of cardiac magnetic resonance in left ventricle noncompaction: a prospective multicenter study. J Am Coll Cardiol. 2016;68(20):2166–81. [DOI] [PubMed] [Google Scholar]
- 7.Grigoratos C, Barison A, Ivanov A, et al. Meta-Analysis of the prognostic role of late gadolinium enhancement and global systolic impairment in left ventricular noncompaction. JACC Cardiovasc Imaging. 2019;12(11 Pt 1):2141–51. [DOI] [PubMed] [Google Scholar]
- 8.Gorter TM, van Veldhuisen DJ, Bauersachs J, et al. Right heart dysfunction and failure in heart failure with preserved ejection fraction: mechanisms and management. Position statement on behalf of the heart failure association of the European society of cardiology. Eur J Heart Fail. 2018;20(1):16–37. [DOI] [PubMed] [Google Scholar]
- 9.Gaynor SL, Maniar HS, Prasad SM, Steendijk P, Moon MR. Reservoir and conduit function of right atrium: impact on right ventricular filling and cardiac output. Am J Physiol Heart Circ Physiol. 2005;288(5):H2140–5. [DOI] [PubMed] [Google Scholar]
- 10.Aquaro GD, De Luca A, Cappelletto C, et al. Prognostic value of magnetic resonance phenotype in patients with arrhythmogenic right ventricular cardiomyopathy. J Am Coll Cardiol. 2020;75(22):2753–65. [DOI] [PubMed] [Google Scholar]
- 11.Weng Z, Yao J, Chan RH, et al. Prognostic value of LGE-CMR in HCM: a meta-analysis. JACC Cardiovasc Imaging. 2016;9(12):1392–402. [DOI] [PubMed] [Google Scholar]
- 12.Mandawat A, Chattranukulchai P, Mandawat A, et al. Progression of myocardial fibrosis in nonischemic DCM and association with mortality and heart failure outcomes. JACC Cardiovasc Imaging. 2021;14(7):1338–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Li G, Zhang Z, Gao Y, et al. Age- and sex-specific reference values of biventricular strain and strain rate derived from a large cohort of healthy Chinese adults: a cardiovascular magnetic resonance feature tracking study. J Cardiovasc Magn Reson. 2022;24(1):63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Voigt J-U, Cvijic M. 2- and 3-dimensional myocardial strain in cardiac health and disease. JACC Cardiovasc Imaging. 2019;12(9):1849–63. [DOI] [PubMed] [Google Scholar]
- 15.Jain S, Kuriakose D, Edelstein I, et al. Right atrial phasic function in heart failure with preserved and reduced ejection fraction. JACC Cardiovasc Imaging. 2019;12(8 Pt 1):1460–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zghaib T, Bourfiss M, van der Heijden JF, et al. Atrial dysfunction in arrhythmogenic right ventricular cardiomyopathy. Circ Cardiovasc Imaging. 2018;11(9):e007344. [DOI] [PubMed] [Google Scholar]
- 17.Gao Y, Pu C, Li Q, et al. Assessment of right atrial function measured with cardiac MRI feature tracking for predicting outcomes in patients with dilated cardiomyopathy. Radiology. 2024;310(3):e232388. [DOI] [PubMed] [Google Scholar]
- 18.Jenni R, Oechslin E, Schneider J, Attenhofer Jost C, Kaufmann PA. Echocardiographic and pathoanatomical characteristics of isolated left ventricular non-compaction: a step towards classification as a distinct cardiomyopathy. Heart. 2001;86(6):666–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Petersen SE, Selvanayagam JB, Wiesmann F, et al. Left ventricular non-compaction: insights from cardiovascular magnetic resonance imaging. J Am Coll Cardiol. 2005;46(1):101–5. [DOI] [PubMed] [Google Scholar]
- 20.Jacquier A, Thuny F, Jop B, et al. Measurement of trabeculated left ventricular mass using cardiac magnetic resonance imaging in the diagnosis of left ventricular non-compaction. Eur Heart J. 2010;31(9):1098–104. [DOI] [PubMed] [Google Scholar]
- 21.Kramer CM, Barkhausen J, Flamm SD, Kim RJ, Nagel E. Standardized cardiovascular magnetic resonance (CMR) protocols 2013 update. J Cardiovasc Magn Reson. 2013;15(1):91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Xue H, Kellman P, Larocca G, Arai AE, Hansen MS. High spatial and temporal resolution retrospective cine cardiovascular magnetic resonance from shortened free breathing real-time acquisitions. J Cardiovasc Magn Reson. 2013;15(1):102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kellman P, Larson AC, Hsu L-Y, et al. Motion-corrected free-breathing delayed enhancement imaging of myocardial infarction. Magn Reson Med. 2005;53(1):194–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kowallick JT, Morton G, Lamata P, et al. Quantification of atrial dynamics using cardiovascular magnetic resonance: inter-study reproducibility. J Cardiovasc Magn Reson. 2015;17(1):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Casas G, Limeres J, Oristrell G, et al. Clinical risk prediction in patients with left ventricular myocardial noncompaction. J Am Coll Cardiol. 2021;78(7):643–62. [DOI] [PubMed] [Google Scholar]
- 26.Han P-L, Shen M-T, Jiang Y, Jiang Z-K, Li K, Yang Z-G. Prognostic value of left atrial reservoir strain in left ventricular myocardial noncompaction: a 3.0 T cardiac magnetic resonance feature tracking study. J Magn Reson Imaging: JMRI. 2023;57(2):559–75. [DOI] [PubMed] [Google Scholar]
- 27.Stacey RB, Andersen M, Haag J, et al. Right ventricular morphology and systolic function in left ventricular noncompaction cardiomyopathy. Am J Cardiol. 2014;113(6):1018–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Leung SW, Elayi C-S, Charnigo RJ, Syed MA. Clinical significance of right ventricular dysfunction in left ventricular non-compaction cardiomyopathy. Int J Cardiovasc Imaging. 2012;28(5):1123–31. [DOI] [PubMed] [Google Scholar]
- 29.Wang W, Chen W, Lin X, Fang L. Influence of right ventricular dysfunction on outcomes of left ventricular non-compaction cardiomyopathy. Front Cardiovasc Med. 2022;9:816404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zheng J-Y, Chen B-H, Wu R, et al. Prognostic value of right atrial strains in arrhythmogenic right ventricular cardiomyopathy. Insights into Imaging. 2024;15(1):58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Schuster A, Backhaus SJ, Stiermaier T et al. Impact of right atrial physiology on heart failure and adverse events after myocardial infarction. J Clin Med 2020;9(1). [DOI] [PMC free article] [PubMed]
- 32.Querejeta Roca G, Campbell P, Claggett B, Solomon SD, Shah AM. Right atrial function in pulmonary arterial hypertension. Circul Cardiovasc Imaging. 2015;8(11):e003521. discussion e003521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Willens HJ, Fertel DP, Qin J, Labrador E, Lowery MH. Effects of age and pulmonary arterial hypertension on the different phases of right atrial function. Int J Cardiovasc Imaging. 2008;24(7):703–10. [DOI] [PubMed] [Google Scholar]
- 34.Goetze JP, Bruneau BG, Ramos HR, Ogawa T, de Bold MK, de Bold AJ. Cardiac natriuretic peptides. Nat Rev Cardiol. 2020;17(11):698–717. [DOI] [PubMed] [Google Scholar]
- 35.Hopman LHGA, Visch JE, Bhagirath P, et al. Right atrial function and fibrosis in relation to successful atrial fibrillation ablation. Eur Heart J Cardiovasc Imaging. 2023;24(3):336–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Goette A, Kalman JM, Aguinaga L et al. EHRA/HRS/APHRS/SOLAECE expert consensus on atrial cardiomyopathies: definition, characterization, and clinical implication. Eur Eur Pacing Arrhythm Card Electrophysiol J Work Groups Card Pacing Arrhythm Card Cell Electrophysiol Eur Soc Cardiol. 2016;18(10):1455–1490. [DOI] [PMC free article] [PubMed]
- 37.Hirono K, Ichida F. Left ventricular noncompaction: a disorder with genotypic and phenotypic heterogeneity-a narrative review. Cardiovasc Diagn Ther. 2022;12(4):495–515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ivanov A, Dabiesingh DS, Bhumireddy GP et al. Prevalence and prognostic significance of left ventricular noncompaction in patients referred for cardiac magnetic resonance imaging. Circul Cardiovasc Imaging 2017;10(9). [DOI] [PubMed]
- 39.Nucifora G, Aquaro GD, Pingitore A, Masci PG, Lombardi M. Myocardial fibrosis in isolated left ventricular non-compaction and its relation to disease severity. Eur J Heart Fail. 2011;13(2):170–6. [DOI] [PubMed] [Google Scholar]
- 40.Mavrogeni S, Sfendouraki E, Theodorakis G, Kolovou G. Diagnosis, severity grading and prognosis of left ventricular non-compaction using cardiovascular magnetic resonance. Int J Cardiol. 2013;167(2):598–9. [DOI] [PubMed] [Google Scholar]
- 41.Zeppenfeld K, Tfelt-Hansen J, de Riva M, et al. 2022 ESC guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death. Eur Heart J. 2022;43(40):3997–4126. [DOI] [PubMed] [Google Scholar]
- 42.Yu S, Chen X, Yang K, et al. Correlation between left ventricular fractal dimension and impaired strain assessed by cardiac MRI feature tracking in patients with left ventricular noncompaction and normal left ventricular ejection fraction. Eur Radiol. 2022;32(4):2594–603. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets analyzed during the current study are available from the corresponding author upon reasonable request.